Few comparable attempts to explain cross-country structural reform patterns have been
made in the literature, owing mainly to data constraints. One exception is IMF (2004), which
seeks to explain the policy stance in a large number of structural policy areas across many OECD
countries between the mid-1970s and the late 1990s. The study explores the impact of a range of
factors including macroeconomic conditions, political institutions, reform design and variables
aimed to capture attitudes towards structural reform. As expected, fiscal surpluses are found to
favour reforms in product and labour markets, while fiscal adjustment -measured as an
improvement in the fiscal position- hinders them.
The remainder of this section proceeds in three steps. Section 4.1 discusses briefly the
expected effects of the explanatory variables featured in the regressions. Section 4.2 presents the
annual dataset of major labour and product reforms assembled for 21 OECD countries over the
period 1985-2003.88 Section 4.3 attempts to explain the probability of carrying out major reforms
by means of simple Probit estimates, and Section 4.4 presents linear estimates as a robustness
check.
4.1 Potential influences on the propensity to undertake reforms
The analysis essentially seeks to determine whether fiscal surpluses and/or fiscal
adjustment facilitate or impede the reform process. As already noted, both variables are captured
by the cyclically-adjusted fiscal surplus and the change in the cyclically-adjusted primary balance,
respectively. Other potential influences on reform patterns which are controlled for in the
econometric analysis are the following: 89, 90
- Macroeconomic situation:
Strict regulation of labour and - possibly -product markets is more likely to be perceived
as counter-productive when labour market performance is weak. Over and above the
possible impact of initial conditions on reform intensity, of which some simple bivariate
evidence was found above for non-EMU countries, economic crises could play a specific
role in fostering reforms. This is because a crisis situation may enable governments to
88.The countries covered: are Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Japan, Netherlands,
New Zealand, Norway, Portugal, Spain, Sweden, Switzerland, United Kingdom, United States.
89.Various attempts to capture the effects of the electoral process on reform patterns were also made but proved unsuccessful. Insofar as structural
reforms have short-term political and/or economic costs, governments may be inclined to postpone them until after general elections have been
held. Therefore, reforms should occur more frequently right after general elections than just before. In order to test for this assumption, a general
election year dummy variable was constructed, taking value 1 on parliamentary or presidential election years and 0 otherwise (Source:
Worldbank, Database on Political Institutions). However, this dummy variable -as well as various lags and leads of it- was always found to be
insignificant in the Probit regressions below.
90.Section 2 also hinted at possible interactions between monetary policy autonomy and the fiscal position in facilitating or hindering the reform
process. The rationale behind such interactions is that independent monetary policy could more easily crowd in resources in the wake of
structural reforms, thereby lessening the need for fiscal accommodation, and vice versa. A dummy variable capturing the degree of autonomy of
national monetary policy was constructed, which took value 1 when the country was not engaged in any kind of fixed exchange-rate arrangement
(e.g. a peg, a monetary union or the former exchange-rate mechanism of the European Monetary System) and zero otherwise (for details, see
Duval and Elmeskov, 2005). However, interactions between this measure of monetary policy autonomy and the fiscal variables consistently
turned out to be statistically insignificant at conventional levels.
189